Hi,
we can now reproduce the problem, but it looks like the problems are
not caused by the kblockd.
The problem on this host is that from time to time "ps -aux" just hangs
and starts eating up all memory. When it has taken enough, either some
kblockd message occurs, and/or the oom killer jumps in and starts killing
threads.
So, the ps -aux hangs *before* the kblockd messages occur, and is not
caused by it (like I assumed before). And since I don't get any disk
errors etc. after the kblockd messages, I guess everything is fine and
the scsi operation indeed recovers the way you said it should.
Now we just need to find out why ps -aux hangs. Seems to be a problem
with the nfsd, because it hangs when showing the [nfsd] entries and
works after restarting the nfs server. In case someone is interested
in this issue, I described it in more detail on the nfs list at
http://marc.theaimsgroup.com/?l=linux-nfs&m=110509676609987&w=2
Thanks for your help!
cu,
Frank
--
Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/
Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/
LMU, Amalienstr. 17 Phone: +49 89 2180-4049
80333 Muenchen, Germany Fax: +49 89 2180-99-4049
* Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *